Grain number determination in an old and a modern Mediterranean wheat as affected by pre-anthesis shading

As the number of grains per unit area has been the main bread wheat (Triticum aestivum) yield determinant affected by breeding, and this variable is highly responsive to pre-anthesis shading, the analysis of the effects of shading on old and modern wheats appears to be a good method to explore how breeding has improved yield. Two field experiments were carried out in a region of Mediterranean Spain with an old cultivar (Aragon 03) and a modern line (advanced line ID-2151) and with 4 shading treatments that reduced by c. 75% the daily incoming solar radiation: unshaded control, shading from jointing to anthesis, shading from jointing to the beginning of booting, and shading from the beginning of booting to anthesis. Grain number per m2 differed between cultivars and was affected by shading. The old cultivar had much fewer grains per spikelet (c. 45%) and slightly fewer spikelets per spike (c. 15%) than the modern line and shading decreased only the number of grains per spikelet in direct proportion to the reduction of the incoming solar radiation, leaving the number of spikes per m2 unaffected by either of the 2 factors. Shading the crop during the entire spike growing period also decreased harvest index (c. 17%), although the reduction in partitioning was not evident at anthesis. As the biomass and spike dry weight at anthesis were similarly and consistently reduced by shading, it seemed that pre-anthesis shading reduced the number of developing florets, with its effects on the number of grains per m2, without major changes in resource allocation among structural components of the spike (rachis, glumes, and awns).